NIMROD modeling of dispersive shell pellet injection is being pursued to evaluate this method as an alternative to shattered pellet injection that may have advantages for both thermal quench and runaway electron mitigation. DSP aims to mitigate disruptions by cooling the plasma from the inside out with a boron payload delivered to the core inside a slowly-ablating diamond shell. DIII-D DSP experiments showed several significant trends as pellet velocity was varied, and which are reproduced in the simulation, helping to validate the model and gain new physics insight into the experiment. In the modeling, as pellet velocity increased, thermal quench mitigation improved and the plasma current spike amplitude decreased, both in agreement with experiment. The plasma current spike is associated with a reconnection event at the x-point, reducing the closed flux volume and producing a large halo-current region (with 70 kA of halo current). It was shown that the faster pellets quickly cool the core to just a few eV and dissipate significant poloidal flux at the o-point, while the slower pellets cool the core less rapidly, producing a flatter temperature profile, and cause poloidal flux to diffuse outward and finally be expelled at the x-point by reconnection with the private flux region. Another experimental finding, that runaway electrons were produced only for the fastest pellet velocity, is also shown to be fully consistent with the modeling, in which rapid cooling of the core to a regime where the electric field exceeds the critical field (satisfying the condition for hot-tail runaway production) occurs only for the deepest penetrating pellet. The simulations also provide insight into the requirements for a predictive model for DSP that can be extended to other tokamaks.

The asymmetry between deuterium (D) and tritium (T) turbulent particle fluxes in mixed D-T plasmas has been studied with numerical simulations of nonlinear gyrokinetic turbulence in ion temperature gradient (ITG)-dominated and trapped electron mode (TEM)-dominated regimes. At 50-50 D-T concentration, an asymmetry (or “flow separation”) develops between the D and T fluxes, implying that the concentration would evolve to an unequal state, or a build-up of one species relative to the other. The asymmetry is such that the tritium is better confined than the deuterium in both ITG and TEM regimes. To supplement the nonlinear simulations, an analytic quasilinear theory of the particle flux symmetry breaking was developed. The theory is valid for general electron dynamics, and correctly predicts the numerically-computed deviation in the ion density fraction (from 50-50), or ion density gradient (from the electron gradient), required to restore equal deuterium and tritium fluxes.

An OMFIT tutorial was held Wednesday February 10 on the OMFIT Stability, Transport, Equilibrium, Pedestal (STEP) module. STEP uses a centralized data structure via OMAS, which allows for the steps to be interchangeable, permitting a variety of workflows. The STEP workflow shown at the tutorial iterates between the TGYRO code to predict kinetic profiles, the EPED code to predict the pedestal, the ONETWO code for current profile evolution, and the EFIT or CHEASE codes for ensuring Grad-Shafranov equilibrium to predict coupled core-pedestal plasmas. Two examples of this workflow were shown. The first example showed the predict-first capabilities of STEP predicting a DIII-D plasma, then modifying neutral beam injections heating and magnetic field to predict how the scenario would change. The second example showed how use STEP to predict the fusion power of a reactor given only zero dimensional parameters.

The ITER Integrated Modeling and Analysis Suite (IMAS) ontology defines a standard for organizing both experimental and simulation data of tokamak experiments. OMAS [https://gafusion.github.io/omas] is an open source Python library designed to facilitate the manipulation of data consistent with the IMAS ontology. OMAS has been extended to enable the on-the-fly translation of experimental data of existing devices to IMAS. The method has been mainly applied to DIII-D, for which an ever-growing subset of experimental signals can be queried following the IMAS notation. Support for other devices (NSTX and EAST) has also been included, and the ability to access data, independently of how their native databases are organized, has been demonstrated. The new capability marks an important cornerstone towards a wider adoption in the US of the standard backed by ITER, as it addresses one of its biggest hurdles: the availability of experimental data in IMAS format.